Evaluation of protective efficacy induced by different heterologous prime-boost strategies encoding triosephosphate isomerase against Schistosoma japonicum in mice

Background In China, schistosomiasis japonica is a predominant zoonotic disease, and animal reservoir hosts in the environment largely sustain infections. The development of transmission-blocking veterinary vaccines is urgently needed for the prevention and efficient control of schistosomiasis. Heterologous prime-boost strategy is more effective than traditional vaccination and homologous prime-boost strategies against multiple pathogens infection. In the present study, to further improve protective efficacy, we immunized mice with three types of heterologous prime-boost combinations based on our previously constructed vaccines that encode triosphate isomerase of Schistosoma japonicum, tested the specific immune responses, and evaluated the protective efficacy through challenge infection in mice. Methods DNA vaccine (pcDNA3.1-SjTPI.opt), adenoviral vectored vaccine (rAdV-SjTPI.opt), and recombinant protein vaccine (rSjTPI) were prepared and three types of heterologous prime-boost combinations, including DNA i.m. priming-rAdV i.m. boosting, rAdV i.m. priming-rAdV s.c. boosting, and rAdV i.m. priming-rSjTPI boosting strategies, were carried out. The specific immune responses and protective efficacies were evaluated in BALB/c mice Results Results show that different immune profiles and various levels of protective efficacy were elicited by using different heterologous prime-boost combinations. A synergistic effect was observed using the DNA i.m. priming-rAdV i.m. boosting strategy; however, its protective efficacy was similar to that of rAdV i.m. immunization. Conversely, an antagonistic effect was generated by using the rAd i.m. priming-s.c. boosting strategy. However, the strategy, with rAdV i.m. priming- rSjTPI s.c. boosting, generated the most optimal protective efficacy and worm or egg reduction rate reaching up to 70% in a mouse model. Conclusions A suitable immunization strategy, rAdV i.m. priming-rSjTPI boosting strategy, was developed, which elicits a high level of protective efficacy against Schistosoma japonicum infection in mice. Electronic supplementary material The online version of this article (doi:10.1186/s13071-017-2036-5) contains supplementary material, which is available to authorized users.


Background
Schistosomiasis is an important neglected tropical disease caused by trematode flatworms of the genus Schistosoma [1,2]. Schistosomiasis transmission has been reported in 78 countries or regions in Africa, Asia and Southern America, and it has been estimated that at least 258.9 million people required preventive treatment in 2014 [3]. In China, schistosomiasis (caused by S. japonicum) is the most severe disease in history. Although extensive achievements have been made through its efficient control in the past several years, schistosomiasis remains endemic in the lowland marsh areas or lake regions of Hunan, Hubei, Jiangxi, Anhui and Jiangsu provinces and in the mountain areas of Sichuan and Yunnan provinces [4,5]. In 2014, it was reported that there were 115,614 cases of schistosomiasis japonica distributed in 453 counties and 919,579 cattle raised in epidemic areas [6].
Praziquantel, an effective chemotherapy drug against S. japonicum that is relatively safe and of low cost, does not prevent host reinfection, and repeated chemotherapy treatment may generate drug resistance or decreased effectiveness against worms [7][8][9][10]. In China, schistosomiasis japonica is also a predominant zoonotic disease, and there are more than 40 animal reservoir hosts in the environment, including water buffalo, cattle, pigs and goats, which in turn largely contribute to sustaining the infection [11,12]. Therefore, development of transmissionblocking veterinary vaccines is urgently needed for the prevention and efficient control of schistosomiasis in China.
Results from seroepidemiological investigation and studies of the radiation-attenuated cercariae model have provided evidence for the feasibility of vaccine development against schistosome infection [13,14]. The World Health Organization (WHO) proposed that a vaccine with partial protective efficacy (≥ 50%) could ease host damage, reduce environmental pollution by eggs, and decrease overall morbidity [15]. Vaccines against S. japonicum have been studied for several years, and numerous antigen candidates from all life stages have been tested, including the 23-kDa membrane protein (Sj23), fatty acid-binding protein (SjFABP), and glutathione-Stransferase (SjGST). However, the protective efficacy induced by these antigens are not as ideal as expected [16][17][18][19]. Therefore, strategies for the improvement of protective efficacy should be further investigated for the development of novel vaccines against S. japonicum infection.
In recent years, a novel vaccination strategy, heterologous prime-boost, which uses unmatched vaccine delivery methods for immunization while using the same antigen, has been extensively applied in vaccine studies and has been determined to be more effective than traditional vaccination strategy of homologous prime-boost strategy [20]. Different prime-boost formats have been widely used in vaccine research against malaria, tuberculosis and AIDS, such as DNA priming-protein boosting and DNA primingviral vectored vaccine boosting [21][22][23]. In our previous study, we cloned and optimized codon usage of the gene, triosephosphate isomerase of S. japonicum (SjTPI) for the first time [24]. Different types of vaccines were constructed, including DNA vaccine (pcDNA3.1-SjTPI, pcDNA3.1-SjTPI.opt), recombinant protein vaccine (rSjTPI), and recombinant adenoviral vaccine (rAdV-SjTPI.opt), and its protective efficacy was evaluated in a mouse model by using homologous prime-boost strategy. The results showed that worm reduction rates did not stabilize at the 50% level, a value recommended by the WHO. However, worm reduction rates significantly increased from 26.9 to 36.9% when a DNA priming-protein boosting strategy was used [17,[24][25][26].
To further improve protective efficacy, the present study immunized mice with three different types of heterologous prime-boost strategies based on our previously constructed vaccines, tested the specific immune responses, and evaluated the protective efficacy through challenge infection of S. japonicum with cercariae.

Animals and parasites
Six-week-old female BALB/c mice were purchased from the Shanghai Laboratory Animal Center (SLAC; Shanghai, China) and used in the vaccination studies. A Chinese mainland strain of S. japonicum infected Oncomelania hupensis was provided by Jiangsu Institute of Parasitic Diseases (Wuxi, China). Cercariae were collected from infected snails and used in animal challenges.

Vaccine preparation
DNA vaccines (pcDNA3.1-SjTPI.opt) were previously constructed and purified by using Qiagen Plasmid Mega Kit (Qiagen, Dusseldorf, Germany) according to the manufacturer's instructions. The final plasmid DNAs were in 0.01 M phosphate buffered solution (PBS) and verified for immunization by restriction enzyme digestion and DNA sequencing [24]. Recombinant proteins (rSjTPI) were purified from a prokaryotic expression system (pGEX-4T-3 as a vector, previously constructed), using a GST-tag purification modules (GE Healthcare; Buckinghamshire, UK), and thrombin (Sigma-Aldrich; St. Louis, USA) was used to remove the GST-tag [27]. The rSjTPI was diluted with PBS to a final concentration of 0.1 mg/ml, stored in aliquots at -80°C and emulsified with an equal volume of Freund's incomplete adjuvant (Sigma-Aldrich; St. Louis, USA) before immunization. Recombinant adenoviral vectored vaccines (rAdV-SjTPI.opt) were constructed and purified previously [26], stored in aliquots at -130°C until use.

Measurement of rSjTPI-specific antibody responses
Serum samples of each mouse were collected from caudal veins before immunization and challenge. Indirect enzyme linked immunosorbent assays (ELISAs) were used to measure rSjTPI-specific antibody responses, including IgG levels, IgG subclass (IgG1 and IgG2a) levels, IgG avidity, and IgG titer. rSjTPI (rTPI, purified previously) was used as the antigen source. To measure IgG, IgG1, and IgG2a levels, serum samples at a 1:100 dilution were added into ELISA plates (Nunc) that were coated with rTPI (0.2 μg/well) and recognized by second antibodies (HRP-conjugated goat-anti-mouse IgG, IgG1, and IgG2a, SouthernBiotech; Birmingham, USA) at a 1:5000 dilution. The optical density (OD) was read at a wavelength of 450 nm with a microplate reader (Antobio; Zhengzhou, China). To assess IgG avidity, an additional washing step with 6 M urea in PBST was performed after serum incubation to discard low avidity IgG, and the avidity index was calculated as the ratio of the OD 450 treated and OD 450 untreated, as described elsewhere [28,29]. To measure IgG titers, serum samples from each mouse were examined using multiple dilutions (from 1:50 to 1:638,400) and the IgG titer was determined by comparing these to the OD 450 value of the control (cut-off value ≥ 2.1 × the mean OD 450 value of the control).

Cytokine measurements
Two days before challenge, four mice from each group were randomly sacrificed, and cell suspensions were prepared under aseptic conditions by grinding the spleens and filtering through 200-mesh screens. The splenocytes from each mouse were cultured in triplicate (cell density: 5 × 10 5 cells per well) in 96-well plates (Corning; NY, USA), incubated in RPMI 1640 medium (Hyclone; South Lagan, USA) supplemented with 10% fetal calf serum (Gibco; Grand Island, USA), and stimulated with rTPI (10 μg/ml), ConA (Sigma-Aldrich; St. Louis, USA, 10 μg/ml), or medium alone (mock) at 37°C with 5% CO 2 for 72 h. The supernatants were collected, and cytokine levels were measured using a BD Cytometric Bead Array (CBA) Mouse Th1/Th2/Th17 Cytokine Kit, according to the manufacturer's protocols.

Elispot assay
Cell suspensions from each group were prepared and stimulated as earlier described. The number of IL-4 and IFN-γ secreting cells were determined using mouse IL-4 and IFN-γ ELISpot kits (R&D; Minneapolis, USA), according to the manufacturer's protocols. Spot forming units (SFU) were counted using the ELISpot Immuno-Spot S5 Analyzer (C.T.L., Germany) and analyzed using the C.T.L. ImmunoSpot image software version 5.1. The results were expressed as SFU for 1 × 10 6 cells.

Detection of specific antibodies against adenoviruses
Viral particles (VPS) of adenoviruses were determined by using the OD 260 method (1 OD 260 = 1.1 × 10 12 VPS/ml) [30]. In addition, indirect ELISAs were performed to detect adenovirus-specific antibody levels. Adenoviruses were used as the antigen source. Serum samples from each group at a 1:100 dilution were added into plates coated with adenovirus (10 7 VPS/well) and recognized by secondary antibodies (HRP-conjugated goat anti-mouse IgG, SouthernBiotech; Birmingham, USA) at a 1:5000 dilution. ODs were read at a wavelength of 450 nm using a microplate reader (Antobio; Zhengzhou, China).

Animal challenge and efficacy observation
Two weeks after the last immunization, each mouse was challenged with 40 ± 1 S. japonicum cercariae by abdominal skin penetration. Forty-two days post-challenge, all mice were sacrificed and perfused to observe worm burdens. Worm (female worm) reduction rate was calculated by using the following formula: Reduction rate (%) = [1 -Average total worm (or female worm) burden in each group/Average total worm (or female worm) burden in the control group] × 100. Whole livers from each mouse were collected, weighted, and digested with 5 ml of 5% potassium hydroxide (KOH) at 37°C for 72 h. Ten microliters of the liver digest were loaded onto a glass counting slide to count the number of eggs (repeated 3 times), and the number of eggs per gram liver from each mouse was calculated. Liver egg reduction rates were calculated by using the following formula: Reduction rate (%) = (1 -Average number of eggs per gram liver in each group/Average number of eggs per gram liver in the control group) × 100.

Histopathological examination of livers
Areas of single egg granuloma in the livers were observed by using sectioned liver tissues (1-5 cm 3 ) collected from each mouse. The procedures of section preparation were according to standard histological operations, including fixation in 4% formaldehyde, dehydration in alcohol, embedding in paraffin, and staining with hematoxylin-eosin. Egg granulomas in the liver were observed and imaged under a light microscope (Olympus BX51; Tokyo, Japan). Areas of each single egg granuloma were determined using a computerized image analysis system (JD801 Version 1.0; Nanjing, China). Granuloma sizes were expressed as the means of areas measured in μm 2 ± SD.

Statistical analysis
Statistical analysis was performed using the SPSS software (Version 19.0). One-way ANOVA was used for data comparison among different groups, and the paired Student's t-test was used to compare any two means. P-values < 0.05 or < 0.01 were considered statistically significant.

Discussion
We evaluated specific immune responses and protective efficacy against S. japonicum in mice using three types of heterologous prime-boost combinations, including DNA i.m. priming-rAdV i.m. boosting, rAdV i.m. priming-rAdV s.c. boosting, and rAdV i.m. priming-rSjTPI boosting strategies. The results of the present study showed that various heterologous prime-boost combinations elicit different immune profiles, and different levels of protective efficacy were generated accordingly. However, the strategy, priming with rAdV intramuscularly, and boosting with rSjTPI subcutaneously, generated the optimal protective efficacy and the worm or egg reduction rate reaching up to 70% in a mouse model.
Previous studies have clearly shown that heterologous prime-boost vaccination elevates protective efficacy [20][21][22][23]. However, its underlying mechanism has not been clearly elucidated. Different vaccine vectors or delivery systems may deliver and present protective antigens in their own way, and this may stimulate the host immune systems to generate antibodies with higher avidity, a broad spectrum of specific immune responses, and the circumvention of anti-vector effects [31,32]. Furthermore, previous studies have shown that a high level of specific Th1 (IFN-γ and IgG2a) responses is associated with a high degree of protection against S. japonicum infection in animal models [33,34]. However, specific Th2 responses may also contribute to protection [35]. In our previous studies, a series of vaccines based on triosephosphate isomerase of S. japonicum (SjTPI) were constructed, including a DNA vaccine (pcDNA3.1-SjTPI.opt), a protein vaccine (rSjTPI), and an adenoviral vectored vaccine (rAdV-SjTPI.opt). Animal experiments have shown that DNA and adenoviral vectored vaccines elicit a specific Th1biased immune response when immunized intramuscularly, whereas, protein and adenoviral vectored vaccines elicit a specific Th2-biased immune response when immunized subcutaneously [19,20,25,26]. To obtain a vaccination Fig. 5 The single-egg granuloma responses in the liver induced by each immunization strategy. a Representative granuloma of each group induced by a single egg in liver (magnification factor 10 × 10; Scale-bars: 100 μm). b Areas of the single-egg granuloma in liver. Data are expressed as the mean ± standard deviation (SD). *P < 0.05; **P < 0.01 Fig. 6 Adenovirus-specific IgG responses by immunized group. Each bar represents the mean ± standard deviation (SD). *P < 0.05; **P < 0.01 strategy with higher efficacy against S. japonicum, we designed three types of heterologous prime-boost strategies in the present study.
DNA and adenoviral vaccines could express the antigen in muscular cells when immunized intramuscularly, which present protective antigens through the MHC-I processing pathway and could elicit specific Th1-biased immune responses [36,37]. Furthermore, the anti-vector effect might be minimized using different vaccine delivery systems (vectors). In the present study, a synergistic effect was produced by using a DNA i.m. priming-rAdV i.m. boosting strategy as indicated by an enhancement of Th1-biased immune responses and protective efficacy compared to that using DNA i.m. immunization. However, the observed protective efficacy was similar to that with rAdV i.m. immunization. This may be due to differences in vivo transfection efficiency between DNA plasmids and adenoviruses immunized intramuscularly because DNA plasmids passively enter cells (penetrate), whereas adenoviruses enter cells through active infection (transfection) [36,37]. This difference may affect the expression of the delivered antigens and ultimately lead to differences in specific immune responses and protective efficacies accordingly.
Replication-deficient adenoviral vectors retain its actively invading ability to target cells and show high transfection efficiency when applied to a vaccine delivery systems [38,39]. Previous studies have shown that different types of immune responses and various levels of protective efficacy are elicited when immunized intramuscularly or subcutaneously [26]. To gain a broad spectrum of specific immune responses, we exploited rAdV-SjTPI.opt immunized intramuscularly as the priming vaccine and used different boosting vaccines (rAdV-SjTPI.opt or rSjTPI, all immunized subcutaneously). Different outcomes were obtained using these two heterologous prime-boost strategies. A synergistic effect was produced by the rAdV i.m. priming-rSjTPI boosting strategy, as indicated by the board spectrum of immune responses and high protective efficacy (> 70%) against infection. However, an antagonistic effect was produced by the rAd i.m. priming-s.c. boosting strategy, as indicated by the moderate levels of immune responses and protective efficacies. Differences in results may be attributable to various in the employed vaccines. As earlier described, adenoviral vectored vaccines immunized intramuscularly can present antigens via the MHC-I way and elicit Th1-biased responses. However, protein or adenoviral vectored vaccines immunized subcutaneously can present antigens via the MHC-II way and elicit Th2biased responses. Anti-vector effects are another group of factors that affect protective efficacy that is elicited by a heterologous prime-boost strategy [31,32,40]. Through the detection of specific anti-adenovirus antibodies, the highest anti-adenovirus antibody levels were observed in the rAd i.m. priming-s.c. boosting group. These specific antibodies could efficiently neutralize adenoviral vectored vaccines as well as cause antagonistic effects on the final outcomes.
Schistosoma japonicum, a genus of complex multicellular pathogen, undergoes six different developmental stages, of which the schistosomulum, adult worm, and egg occur within the definitive host [2]. The specific immune responses against S. japonicum infection are complex and have not been clearly elucidated [13]. Previous studies have shown that Th1-biased immune responses play an important role in protecting against infections in a radiation-attenuated cercariae animal model [14,34]. Furthermore, specific IgG responses may also contribute to an increase in protective efficacy [35]. In the present study, the rAdV i.m. priming-rSjTPI boosting strategy elicited broad spectrum immune responses, which were manifested as higher IgG responses (IgG levels, IgG titers, and IgG avidity), elevated Th1, Th2, and Th17 cytokine levels, as well as produced the highest level of protective efficacy among the three heterologous prime-boost combinations. These results were in agreement with those reported in previous studies.

Conclusion
In summary, we have developed a suitable immunization strategy, rAdV i.m. priming-rSjTPI boosting strategy, which elicits a high level of protective efficacy against S. japonicum infection in mice. However, comparison of different heterologous prime-boost combinations indicated that different factors may be considered when designing a suitable heterologous prime-boost strategy, including types of protective immune responses against infection, characteristics of different vaccines, anti-vector effects, and suitable vaccination routes.